Construction And Research Of Multi-factor Quantitative Model Of A-share Market

In recent years,China's rapid economic development has increased the demand for diversified investment.As an indispensable part of the financial business,the securities market has made great progress since its establishment.From 2014 to the end of 2017,the scale of China's asset management industry jumped from 20 trillion to 60 trillion,expanding threefold.At the same time,the regulatory authorities are actively promoting financial innovation while actively building a sound financial environment.Due to the non-linearity and complexity of the securities market,traditional investment strategies are difficult to meet people's expectations.Therefore,quantitative investment strategies based on computer and big data are gradually favored.Quantitative strategy models are increasingly becoming predictive markets and the mainstream tool to guide investment,and the capital market has an increasing demand for quantitative and efficient investment.Quantitative investment has broad prospects for development in China's securities market.It is still in its infancy,and research on quantitative investment will help China's financial market to further improve.In this context,it is of great significance to study quantitative investment in broadening investment channels,improving market construction and promoting stock market development.Quantitative investment as a process of using computer technology and mathematical models to practice investment ideas is undoubtedly the logic of investing in the data,looking for effective investment opportunities in the market,and actively investing ideas and the behavior of the financial market through a large amount of data and calculations.On the one hand,quantitative investment strategy is composed through computer programs,which avoids personal subjective emotions affecting data analysis.On the other hand,quantitative investment is more real-time and fast,that is,processing data is accurate and efficient,and can effectively,quickly capture valuable information on the market.The quantification process contains four components: financial products,historical data,investment strategies(computer programs),and trading platforms.At present,China's financial transaction products have been relatively rich,and the historical financial data accompanying it has been improved and refined.The quantitative investment strategy has been recognized and accepted by more and more investors with the deepening of academic theoretical research.At the same time,due to the rise of the online quantitative investment platform,the trading threshold has lowered and the platform has become increasingly diverse and open.This paper is based on a brief analysis of the popular network quantitative investment platform-Join Quant quantitative investment platform.The article contains below sections.Based on the current academic literature,research results and practical experience,discusses and reviews the quantitative investment from its historical development,core elements,characteristics analysis and classification strategy.In the second part,the multi-factor quantitative model is used as the research basis.Historical data backtesting and simulation are carried out on the basic multi-factor quantitative strategy to verify the effectiveness of the strategy,and finally to study and demonstrate the theory of quantitative stock picking.Specifically,this paper selects the stocks of the Shanghai and Shenzhen 300 stocks as the basis,and processes the stock data and financial data of the Shanghai and Shenzhen 300 stocks from 2014 to 2017,mainly selecting the following five basic data: valuation,balance,cash flow,income and financial indicator data,from which to determine the range of factors most likely to cause stock price changes,and ultimately through the cumulative excess returns of stock portfolios under each factor,the stability of quantile returns,the probability of defeating the market,the evaluation indicators screen out 8 effective factors for stock selection verification.The eight effective factors are: circulating?market?cap,ps?ratio,eps,net?profit?to?total?revenue,roe,pcf?ratio,pe?ratio,turnover?ratio.At the same time,the invalid redundancy factor is eliminated,and the equal weight method is used to score the factors according to the historical financial data in each historical period,and then the total factor scores are added to obtain the comprehensive score performance of each stock,select the top 20 stocks with the highest scores,and test the historical performance of the portfolio.The whole process of strategy factor score,screening and backtesting process are all carried out on the Join Quant quantitative platform,including historical financial data acquisition(various data of each factor in different time periods),strategy program implementation(based on PYTHON language and Data analysis tools Pandas,Numpy),graphical analysis and indicator presentation,etc.,finally proposed and tested a multi-factor quantitative strategy that can have stable excess returns in the stock market for a long time.In the third part,this paper demonstrates the optimization on this basic strategy,mainly from the three aspects of strategy factor screening,number of positions and frequency of adjustment,and optimizes the strategy by comparing multiple factors in different positions.Base on the historical performance,the optimized strategy shows better annualized income,Alpha,Sharp ratio and other indicators than the basic strategy.Based on the researching of the quantitative investment strategy and application test,this paper come up with the dual purpose of quantitative strategy design,theory verification and accumulates experience.It can be regarded as the initial results of strategic research and this is a good start for personal quantitative investment research in the future.